Can multi-thread accelerate system level simulation which is based on discrete event driven mechanism？

My project is a system level simulation platform which is based on discrete event driven mechanism. As time grows, the platform becomes more and more complex. Unfortunately, the platform is getting slower and slower. To finish a simulation job, the time is getting longer than before. Now, my platform is based on single thread. I want to use multi thread in order to accelerate the platform. Is it possible? If yes, what should I do?